← 返回 Skills 市场
102
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install self-verify
功能描述
Automatically checks and verifies consequential claims before response to prevent confident errors and ensure accuracy and safety.
安全使用建议
This skill appears coherent and low-risk: it contains a clear checklist and suggests checking sources and memory. Before installing, confirm (1) what agent tools 'mmx search' and 'memory_search' actually call in your environment so you know where queries and memory lookups go; (2) that consulting MEMORY.md and any memory APIs doesn't leak sensitive data you don't want included in verification. Because the skill can be invoked autonomously by default, if you prefer manual control, ensure platform settings restrict autonomous invocation or only enable the skill when needed. Test the skill in a safe context first to verify it behaves as expected.
功能分析
Type: OpenClaw Skill
Name: self-verify
Version: 1.0.0
The skill bundle is a meta-cognitive framework designed to improve AI accuracy through self-verification. It provides instructions for the agent to cross-check claims using standard tools like 'mmx search' and 'memory_search' and includes a safety checklist to prevent confident errors. No malicious logic, data exfiltration, or unauthorized execution patterns were found in SKILL.md or _meta.json.
能力评估
Purpose & Capability
Name/description (self-verification of claims) matches the SKILL.md checklist and verification actions. The skill only asks the agent to check sources, memory, uncertainty, contradictions, reversibility, and safety — all appropriate for a self-verification helper.
Instruction Scope
The runtime instructions refer to running cross-checks via 'mmx search' and 'memory_search' and to consulting MEMORY.md and the agent's reasoning trace. These are reasonable verification steps, but they implicitly assume those tools/endpoints exist and are safe; the skill does not declare or constrain which search/memory tool to use. Confirm what 'mmx' and 'memory_search' map to in your agent environment.
Install Mechanism
Instruction-only skill with no install spec, no files written to disk, and no external downloads — minimal install risk.
Credentials
No environment variables, credentials, or config paths are requested. Asking to consult memory/search is proportionate to the stated purpose and does not require additional secrets.
Persistence & Privilege
Skill is not always-enabled and is user-invocable (defaults). It can be invoked autonomously per platform default, which is expected for a helper skill; there is no evidence it requests elevated or persistent privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install self-verify - 安装完成后，直接呼叫该 Skill 的名称或使用
/self-verify触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release: pre-delivery error check - 5-step verification before returning consequential claims
元数据
常见问题
Self Verify 是什么？
Automatically checks and verifies consequential claims before response to prevent confident errors and ensure accuracy and safety. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 102 次。
如何安装 Self Verify？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install self-verify」即可一键安装，无需额外配置。
Self Verify 是免费的吗？
是的，Self Verify 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Self Verify 支持哪些平台？
Self Verify 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Self Verify？
由 ariffazil（@ariffazil）开发并维护，当前版本 v1.0.0。
推荐 Skills